The Psychotomimetic Nature of Dreams: An Experimental Study

Several theories promote the similarities between dreaming and psychosis, but this has rarely been tested empirically. We assessed dreaming and waking reality using the Psychotomimetic States Inventory, a measure of psychotic-like experience originally designed for drug studies. Twenty participants completed the measure in each of two dream conditions and one waking condition. Dreams were assessed upon waking naturally and also using a movement-activated (actigraph) alarm during the night. Overall, participants reported more quasipsychotic characteristics during dreams (in both conditions) than when awake. This was most marked for paranoia and delusional thinking, but differences were also seen for perceptual abnormalities, mania, and anhedonia. The quality of dream experience seems particularly similar to psychosis in sometimes being highly self-referential and having a paranoid content. Subjective changes to cognition and affect are consistent with alterations in prefrontal cortical activity during REM sleep that mirror those of schizophrenia. 1. Introduction and Methods The phenomenological similarity between the dream state and psychosis has been often remarked upon but rarely empirically tested. Kant declared “the lunatic is a wakeful dreamer,” while Schopenhauer said “a dream is a short-lasting psychosis, and a psychosis is a long-lasting dream.” Perhaps the most developed neurobiological account, Hobson [1] has advanced a recent hypothesis of the dreaming brain as a model of psychosis, suggesting dreams mimic the distortion of reality seen in the positive symptoms of schizophrenia, though he prefers the term “delirium.” “In dreams, as in delirium, consciousness is clouded, attention is distractible, intellectual functions are dull, perceptions are hallucinatory, cognition is illogical, emotion is unstable and uncontrolled, memory is poor, and thought processes are at a concrete, not a symbolic, level” [2, page 23]. Although in some ways adversaries of Hobson’s wider model (see [3]), Sohms and Turnbull also advocate that “the functional anatomy of dreaming is almost identical to that of schizophrenic psychosis” [4, page 213]. Pushing the analogy to the furthest by describing psychosis itself as a dream state with the neurobiology of REM stage sleep, Gottesmann [5] outlines the best delineated areas of neurobiological overlap in terms of “common intracerebral disconnections, disturbed responsiveness and sensory deafferentation processes (and) dorsolateral prefrontal deactivation” (page 1105). The commonest metric for assessing and asserting this